Hello everyone!
I am Yang Cheng. I graduated as Computer Science Master's Student at the University of Southern California. I used to work as a Student Research Assistant at USC Information Science Institution, supervised by Dr. Mohamed E Hussein, where I conduct research in adversarial attacks. I believe that researching adversarial examples can be a good way to understand machine learning models. Specifically, could the interpretation of adversarial examples lead to a better understanding of the internal representations and structures of models? If so, could we apply that new understanding to build more interpretable and robust network structures? There is still a long way to go. 😊Recently I started to learn about AI-generated content. I hope to do some research in this new area. I am working with Prof. Sun Lichao from Lehigh Univeristy.
I am actively seeking out Ph.D. opportunities for 2023-2024. Here is my Resume. If you are interested, don't hesitate to contact me through my email below.
ycheng04@x, x=usc.edu
​
​
​
Yang Cheng
ML Researcher and Student @ University of Southern California
My Bachelor's degrees are in Mathematics and Economics. Rather than becoming a financial engineer, I realized I wanted to be a research scientist, which brings me here to you. My interest includes adversarial learning, explainable AI, interpretability, and optimization. Interesting facts: I can give a 3-hour lecture on transformer models before realized.
You can learn more about me here!
​
Interesting Examples from previous project
I deliberately increased the attacking budget. We can tell that attacking different object detection models will give us completely different patterns. One from MTCNN, another from RetinaFace.
Publications
Spontaneous gestures encoded by hand positions can improve language
models: An Information-Theoretic motivated study
Author: Yang Xu, Yang Cheng
Vision-based tracking of library bookshelf for inventory management and smart libraries (Preprint)
Author: Preetham Manjunatha, Yang Cheng, Sucheng Wang, Sami F. Masria, Caroline Muglia
Gestures Are Used Rationally: Information Theoretic Evidence from Neural Sequential Models
Author: Yang Xu, Yang Cheng, Riya Bhatia
Proceedings of the 29th International Conference on Computational Linguistics
Hierarchical data visualization of experimental erythrocyte aggregation employing cross correlation and optical flow applications
Author: Bruce I. Gaynes, Mark B.Shapiro, Abel Saju Augustine, Yang Xu, Yang Lin, Parisa Mirbod, Robert S.Dieterm, Yang Cheng, Mengren Wu, Harish Venkataraman, Yuan Gao, Plamen Petrov, Jie Xu